Pre-treatment liver magnetic resonance elastography and proton density fat fraction predicts adverse outcomes in Asia patients with hepatocellular carcinoma

Abstract

Background: Magnetic resonance elastography (MRE) and liver fat calculation via proton density fat fraction (PDFF) are emerging magnetic resonance techniques for non-invasively assessing liver fibrosis and steatosis respectively. The role of MR-based imaging metrics in the pre-treatment prognostication of hepatocellular carcinoma (HCC) remains unclear. Methods: We prospectively recruited consecutive patients with newly diagnosed HCC referred to the Department of Surgery, Queen Mary Hospital, Hong Kong. Pre-treatment MRE and liver fat PDFF (1.5 Tesla Optima 450W MR scanner, GE) were performed on tumor and non-tumor regions of the liver and analyzed using a MR-sequence software (IDEAL-IQ, GE). An example of MRE and PDFF assessment is depicted in figure 1A and 1B respectively. HCC treatment was categorized as curative (resection, ablation, transplant) or noncurative (locoregional or systemic therapy). Outcomes of HCC recurrence, hepatic decompensation and mortality were assessed. Results: In this interim analysis, 122 patients were recruited (mean age 62.8+/- 8.0 years, 80.3% male). 79.5% (n=97) had chronic hepatitis B, while 70.5% (n=86) and 29.5% (n=36) received curative and noncurative treatment respectively. Median tumor and non-tumor stiffness were 4.3 (3.1-6.0) kPa and 3.4 (2.9-4.5) kPa respectively, while median tumor and non-tumor PDFF were 2.4% (1.2%-5.6%) and 2.6% (1.6%-5.2%) respectively. The median duration of follow-up was 18.6 (12.1-20.9) months; 21 patients (17.2%) died and 28 patients (23.0%) developed decompensation. 26 patients (30.2%) who received curative therapy developed HCC recurrence. In the curative group, via multivariate analysis, increased non-tumor stiffness and non-tumor PDFF were independently associated with HCC recurrence (HR 1.32, 95%CI 1.03-1.71 & HR 1.09, 95%CI 1.02-1.17 respectively, both p<0.05). Increased non-tumor stiffness was independently associated with decompensation (HR 1.38, 95%CI 1.05-1.81, p=0.02), while increased non-tumor PDFF was associated with mortality (HR 1.12, 95%CI 1.01-1.23, p=0.03). In the non-curative group, tumor stiffness was independently associated with decompensation (HR 1.60, 95%CI 1.14-2.24, p<0.01) and no association was noted between imaging metrics and mortality. Conclusion: MRE and PDFF demonstrated a prognosticating role in pre-treatment HCC evaluation. The measurements of both imaging metrics should be considered in the diagnostic workup of HCC.